Why Wall Street Is Divided on the AI Bubble Question

Why Wall Street Is Divided on the AI Bubble Question in 2026

Wall Street AI bubble debate: Wall Street is divided because the same AI boom presents two conflicting narratives: corporate profits and real demand support the bull case. In contrast, extreme capital expenditure, market concentration, and uncertain future returns support the bear case.

The disagreement is not simply about whether artificial intelligence works.

Most major Wall Street institutions agree that AI has become a major technological and economic force.

The argument is about valuation.

How much future profit can the AI industry realistically produce?

How much infrastructure spending can companies sustain?

And are investors paying too much today for earnings that may arrive years from now?

Those questions sit at the center of the Wall Street AI bubble debate.

Torsten Slok of Apollo Global Management has raised concerns about market concentration and the financial structure surrounding AI investment. Jim Covello, head of Global Equity Research at Goldman Sachs Research, has questioned whether AI spending can deliver sufficient returns. Goldman Sachs strategists have also argued that current conditions contain bubble-like features while stopping short of calling the market a full bubble. The Bank for International Settlements has warned that the AI infrastructure boom is becoming increasingly debt-financed.

At the same time, analysts including Jeff Buchbinder of LPL Financial and Michael Wilson of Morgan Stanley have argued that today's AI market differs from the late-1990s dot-com period because leading companies already generate large profits and maintain stronger balance sheets.

The result is a genuine disagreement among analysts.

60-Second Summary

Bull case: Nvidia, Microsoft, Alphabet, Meta and other large technology companies already generate substantial revenue and profits.
Bear case: AI infrastructure spending is rising faster than proven AI software profits.
Goldman Sachs: The firm sees bubble-like elements but has argued that fundamental earnings growth separates the current market from previous speculative bubbles.
Jim Covello: Goldman Sachs' AI skeptic argues that enterprise buyers and AI companies still need to demonstrate stronger financial returns.
Torsten Slok: Apollo's chief economist has warned about market concentration and the financial consequences of extreme capital allocation.
BIS: The Bank for International Settlements says the AI boom increasingly depends on debt financing and high future earnings expectations.
The central disagreement: Bulls believe that current spending builds infrastructure that will deliver decades of productivity gains. Bears believe investment may be outpacing end demand and profits.

Why Is Wall Street Divided on the AI Bubble?

The Wall Street AI bubble debate exists because different analysts focus on different parts of the AI economy.

Bullish analysts focus on revenue growth, corporate profits, computing demand, and productivity gains.

Bearish analysts focus on capital expenditure, market concentration, financing structures,s and uncertain returns on investment.

Both groups can point to real data.

Goldman Sachs Research noted in July 2026 that AI-related companies had added roughly $27 trillion in market value since late 2022.

Goldman also reported that U.S. technology investment as a share of GDP had exceeded its 1990s peak.

That data supports concern about the scale of the investment cycle.

However, Goldman also found major differences from the dot-com era.

Corporate profits have reached new highs, and the balance sheets of leading technology companies remain stronger than those of many speculative internet companies during the late 1990s.

The disagreement therefore comes down to one question.

Will future AI profits justify today's investment?

The Bullish Case Against an AI Bubble

The bullish side of the AI analysts' bullish vs. bearish debate does not deny that AI stocks have risen sharply.

Bullish analysts argue that strong price gains alone do not prove a bubble.

1. Major AI Companies Already Generate Real Profits

One of the biggest differences between the current AI boom and the dot-com era is profitability.

Nvidia sells AI chips to customers with large budgets.

Microsoft generates revenue from cloud computing and enterprise software.

Alphabet earns large profits from advertising and cloud services.

Meta funds AI investment from its existing advertising business.

Amazon generates substantial cash flow from AWS and its broader commerce operations.

These companies are not early-stage startups depending entirely on future promises.

Goldman Sachs Research cited the strength of corporate profits and balance sheets as one reason the current market does not yet match the structure of previous bubbles.

2. Jeff Buchbinder Sees a Smaller Valuation Extreme Than 2000

Jeff Buchbinder of LPL Financial has argued that the AI market does not yet resemble the final stage of the dot-com bubble.

His comparison focuses on valuation and market performance.

The Nasdaq-100 had risen sharply since the launch of ChatGPT, but the increase remained far below the extraordinary gains of the late 1990s.

Technology valuations also remained below the levels reached near the March 2000 market peak.

The argument does not claim that valuations are cheap.

It argues that the market has not reached the same level of speculative pricing seen at the peak of the dot-com era.

3. Michael Wilson Sees an Earnings Story

Morgan Stanley Chief U.S. Equity Strategist Michael Wilson has taken another position.

Wilson has argued that the market should not be viewed only through valuation multiples.

His bullish interpretation focuses on earnings growth.

AI infrastructure spending increases revenue for semiconductor companies, networking suppliers, power equipment manufacturers, and cloud providers.

Wilson's position is that the market can sustain higher prices when earnings rise as well.

This creates an important difference between two types of market rallies.

A speculative rally depends mainly on investors paying higher valuation multiples.

An earnings-driven rally occurs when companies produce higher profits.

The current debate centers on whether AI earnings can continue expanding after the infrastructure build-out slows.

4. The AI Adoption Cycle May Still Be Early

Bullish analysts also argue that the largest wave of AI adoption has not yet happened.

Many companies are still testing AI tools.

Enterprise adoption can take years.

Businesses must integrate AI into software systems, security procedures and daily operations.

If AI eventually increases productivity across finance, healthcare, manufacturing, software development, and logistics, current infrastructure spending could yield long-term returns.

This argument supports the AI bubble opinions of analysts who see the current period as an infrastructure build-out rather than a speculative peak.

The Bearish Case for an AI Bubble

The bearish argument focuses less on whether AI technology has practical uses.

The main concern is whether financial expectations have moved too far ahead of proven demand.

1. Capital Expenditure Is Rising Extremely Fast

Goldman Sachs Research estimated that AI-related investment could exceed $1 trillion in 2026 under a broad global measurement.

Large technology companies continue expanding data centers and computing capacity.

That spending creates a large financial hurdle.

Future revenue must eventually cover the cost of chips, buildings, electricity, networking,g and financing.

Bears ask whether end customers will spend enough money on AI products to support that infrastructure.

2. Jim Covello Questions the Economic Returns

Jim Covello, Head of Global Equity Research at Goldman Sachs, has become one of Wall Street's most visible skeptics regarding AI economics.

Covello does not argue that AI technology is useless.

His concern focuses on financial returns.

In June 2026, Covello said enterprise buyers, AI model companies and hyperscalers had not yet demonstrated enough returns relative to the scale of investment.

His argument is straightforward.

Businesses can spend hundreds of billions of dollars on infrastructure, but investment eventually needs to produce profits.

Covello has questioned whether current economic conditions have moved closer to that result over the past several years.

This creates one of the clearest bearish arguments in the Wall Street AI bubble debate.

The issue is not technological capability.

The issue is return on invested capital.

3. Market Concentration Has Increased

Torsten Slok, Chief Economist at Apollo Global Management, has repeatedly examined concentration in the U.S. stock market.

His analysis has focused on the growing influence of a small group of very large technology companies.

When a small number of companies account for a large share of index performance, investors become more exposed to the same investment thesis.

This concentration can increase market risk.

If expectations for AI spending weaken, the same companies that pushed the market higher can also create pressure on major indexes.

Slok's concern is therefore broader than the valuation of one company.

It concerns the concentration of capital across equities, private markets and infrastructure investment.

4. AI Financing Is Increasingly Connected to Debt

The Bank for International Settlements has added another concern to the debate.

AI infrastructure spending is becoming increasingly debt-financed.

In January 2026, BIS economists Iñaki Aldasoro, Sebastian Doerr and Daniel Rees wrote that the scale of future AI investment would require companies to move from financing through operating cash flow toward greater use of debt.

Private credit has also become more involved.

In March 2026, BIS researchers reported that hyperscaler corporate bond issuance exceeded $100 billion during 2025.

Debt does not automatically create a bubble.

However, debt increases financial pressure because interest and principal obligations remain even if AI revenue slows.

Goldman Sachs: Bubble-Like Features but Real Earnings

Goldman Sachs provides one of the most balanced institutional views among analysts in the AI bubble debate.

In October 2025, Goldman Sachs Research identified several characteristics that resembled previous market bubbles.

These included rising valuations, high market concentration, increased capital intensity, and vendor financing.

Those are real warning signs.

However, Goldman Sachs also identified important differences.

Fundamental Growth vs Pure Speculation

Goldman argued that technology stock appreciation had been supported by actual earnings growth.

This differs from periods when valuations increased without comparable improvements in corporate profits.

Strong Corporate Balance Sheets

Leading AI companies generally have large cash balances and strong existing businesses.

Microsoft, Alphabet, Meta and Amazon can fund substantial investment through operating cash flow.

This financial position differs from the debt-heavy or unprofitable companies that became symbols of previous speculative periods.

Limited Competition at the Top

Goldman Sachs also noted that many historic bubbles involved a flood of new companies competing for investor capital.

The AI infrastructure market has remained concentrated among a smaller group of established companies.

That concentration creates its own risks, but it differs from the mass IPO speculation seen during the dot-com period.

Jim Covello's Profitability Warning

Jim Covello's analysis focuses on a question that every investor eventually faces.

How much money will the investment generate?

AI infrastructure companies have already received substantial revenue from the capital spending cycle.

Nvidia and other suppliers benefit when hyperscalers purchase chips and equipment.

The harder question concerns downstream demand.

Will businesses and consumers eventually spend enough on AI products to justify the infrastructure?

Q: Why does Jim Covello question AI profitability?

A: Jim Covello argues that the AI industry has spent enormous amounts on computing infrastructure without yet proving that enterprise customers, AI model companies, and hyperscalers can generate returns proportional to that investment. His concern focuses on the gap between capital expenditure and proven end demand.

Covello's position does not require AI adoption to fail.

AI can become widely adopted even as investment returns remain disappointing for some companies.

This distinction is central to the financial analysis.

A useful technology does not guarantee that every company investing in the technology will earn an attractive return.

Torsten Slok and the Concentration Risk

Torsten Slok is the Chief Economist at Apollo Global Management.

His work frequently examines U.S. economic conditions, financial markets, interest rates, and market concentration.

Slok's AI-related warnings focus heavily on the amount of capital flowing into a small group of technology companies and related infrastructure.

His concern is connected to diversification.

When AI companies dominate equity returns, venture capital investment,t and infrastructure spending, many parts of the financial system become dependent on similar assumptions about future AI demand.

Concentration Can Amplify Market Moves

Concentration does not cause a market crash by itself.

It can increase the effect of negative surprises.

If a large group of investors owns the same AI companies and those companies disappoint investors, selling pressure can spread quickly.

This explains why Slok's analysis matters to the Wall Street AI bubble debate.

His argument is not limited to one AI stock.

It examines how capital allocation has become increasingly concentrated around a common technological theme.

BIS: Debt Financing Changes the Risk

The Bank for International Settlements has provided one of the most detailed institutional analyses of AI financing.

In January 2026, BIS economists stated that AI investment was rising both in nominal terms and as a share of GDP.

They also noted that future investment needs could require a shift toward debt financing.

By March 2026, BIS research showed that major technology companies had increased corporate bond issuance to support multi-year AI infrastructure projects.

Why Debt Matters

Equity financing absorbs losses differently from debt.

Shareholders can experience falling stock prices without creating a mandatory repayment schedule.

Debt requires interest payments and eventual repayment.

If AI infrastructure generates strong revenue, borrowing can be manageable.

If revenue disappoints, debt can increase financial pressure.

The BIS has described current macroeconomic and financial stability risks as moderate while also warning that the sustainability of the boom depends heavily on companies meeting high earnings expectations.

Bullish vs Bearish AI Analyst Views

Institution or Analyst	General Position	Main Argument	Main Data Focus
Goldman Sachs Research	Balanced	Bubble-like features exist, but strong profits and balance sheets differentiate the current cycle from past bubbles.	Valuations, earnings, market concentration
Jim Covello, Goldman Sachs	Bearish on current economics	AI investment must eventually produce stronger returns for enterprise buyers and infrastructure investors.	Profitability and return on investment
Torsten Slok, Apollo Global Management	Cautious	Market concentration and capital concentration increase exposure to the AI investment thesis.	Index concentration and capital allocation
Bank for International Settlements	Institutionally cautious	AI investment increasingly relies on debt and private financing while future earnings remain uncertain.	Debt issuance and financial stability
Jeff Buchbinder, LPL Financial	Bullish relative to dot-com comparison	Current valuations and market gains remain below late-1990s extremes.	Forward earnings multiples and market performance
Michael Wilson, Morgan Stanley	Bullish	The AI-driven market can remain supported if earnings continue growing.	Corporate earnings growth

Why Analysts Reach Different Conclusions

The conflicting forecasts are not necessarily the result of analysts ignoring the same information.

They often use different time horizons.

Short-Term Analysts Focus on Valuation

Short-term market analysts may focus on whether stock prices already reflect unrealistic expectations.

Long-Term Analysts Focus on Productivity

Long-term analysts may believe AI will create large economic benefits over the next decade.

Credit Analysts Focus on Financing Risk

Credit analysts examine debt, interest costs, and borrowers' ability to repay financing.

Equity Analysts Focus on Earnings Growth

Equity analysts may place greater weight on revenue, margins,s and future earnings estimates.

This creates different conclusions from the same AI investment cycle.

The Numbers Investors Should Watch

Instead of trying to predict exactly when an AI bubble will form or burst, investors can monitor measurable financial data.

1. AI Capital Expenditure: Monitor whether hyperscaler spending continues to rise and whether revenue grows at a similar pace.

2. Free Cash Flow

High capital expenditure can reduce free cash flow even when revenue continues increasing.

3. AI Software Revenue

Infrastructure spending requires end customers who will pay for AI services.

4. Debt Issuance

Increasing corporate borrowing can indicate that investment is moving beyond internally generated cash flow.

5. Data-Center Utilization

New infrastructure must eventually generate enough revenue to justify construction costs.

6. Profit Margins

Investors should compare revenue growth with operating margins and infrastructure costs.

7. Market Concentration

A market dominated by a small number of AI companies can experience larger swings when expectations change.

8. Enterprise Adoption

The long-term AI investment case depends heavily on businesses paying for AI products after experimentation moves into regular operations.

Investor Review Checklist for the AI Bubble Debate

Check AI revenue: Determine whether growth comes directly from AI products or from overall business growth.
Compare capex with cash flow: Rising investment is easier to sustain when operating cash flow increases as well.
Review debt issuance: Identify whether companies increasingly depend on borrowing.
Check valuation multiples: Compare current valuations with expected earnings growth.
Measure market concentration: Track how much of major index performance comes from AI-related companies.
Follow enterprise adoption: Watch whether businesses move from AI experiments to recurring spending.
Monitor margins: Revenue growth without sustainable margins can weaken the investment case.
Review vendor financing: Examine whether suppliers provide financing or guarantees that increase financial connections between buyers and sellers.
Study credit markets: Rising borrowing costs can affect data-center and infrastructure projects.
Separate technology from valuation: AI can succeed as a technology even if some AI stocks become overpriced.

Technical Glossary

1. CAPEX

Capital Expenditure. Money spent on long-term physical assets such as data centers, servers, and AI chips.

2. FCF

Free Cash Flow. Cash remaining after a company pays operating expenses and capital expenditure.

3. ROI

Return on Investment. A measurement of the financial benefit generated relative to the money invested.

4. GPU

Graphics Processing Unit. A specialized processor widely used to train and operate artificial intelligence systems.

5. BIS

Bank for International Settlements. An international financial institution that conducts research on banking, credit markets,s and global financial stability.

Frequently Asked Questions

1. Why is Wall Street divided on the AI bubble?

Wall Street analysts disagree because the AI market contains both strong fundamentals and clear financial risks. Major technology companies generate substantial profits, but AI capital expenditure has also reached extremely high levels. Bullish analysts expect future productivity gains and revenue growth. Bearish analysts question whether those future returns can justify current spending and valuations.

2. Does Goldman Sachs think AI is a bubble?

Goldman Sachs has presented a mixed view. Its research has identified rising valuations, market concentration, capital intensity, ty and vendor financing as bubble-like features. However, Goldman has also argued that strong earnings growth and healthy corporate balance sheets distinguish the current AI cycle from earlier speculative bubbles.

3. What is Jim Covello's view on AI?

Jim Covello of Goldman Sachs has focused on the economic return from AI investment. His concern is that enterprise buyers, AI model companies, and hyperscalers have spent enormous amounts on infrastructure without yet demonstrating returns commensurate with the scale of the investment.

4. What has Torsten Slok said about AI market risk?

Torsten Slok, Chief Economist at Apollo Global Management, has focused on market concentration and the amount of capital flowing toward a small group of AI-related companies. His analysis raises concerns about investors becoming heavily exposed to the same technology and infrastructure thesis.

5. What does the BIS say about the AI boom?

The Bank for International Settlements has reported that AI investment is increasingly moving toward debt financing. BIS researchers have warned that the sustainability of the investment cycle depends on companies meeting high future earnings expectations while borrowing and private credit play larger roles in financing infrastructure.

6. How is the AI boom different from the dot-com bubble?

One major difference is profitability. Many leading AI companies already generate substantial revenue and profits. Companies such as Nvidia, Microsoft, Alphabet, Meta, and Amazon have large existing businesses and strong balance sheets. During the dot-com era, many highly valued companies had limited revenue and weak profits.

7. What could cause AI stocks to fall sharply?

AI stocks could face pressure if hyperscaler capital expenditure slows, AI software revenue disappoints, corporate debt growth outpaces cash flow, or investors reduce valuation multiples. A market decline could occur even if AI technology continues improving.

8. Are AI valuations currently higher than dot-com valuations?

Some technology valuations are high, but several analysts have noted that broad market valuation measures remain below the extremes reached during the late 1990s. The comparison depends on the company, the valuation metric, and the period being measured.

9. Is AI infrastructure spending profitable?

Some parts of the AI infrastructure supply chain are already highly profitable. Semiconductor companies and cloud providers have received substantial revenue from AI spending. The larger unanswered question concerns whether end customers will generate enough economic value from AI applications to sustain the current level of infrastructure investment.

10. What should investors monitor in the AI bubble debate?

Investors should track capital expenditures, AI-related revenue, free cash flow, debt issuance, corporate profit margins, enterprise adoption,n and stock valuations. These measurements provide more useful evidence than headlines alone.

Final Review Framework

The Wall Street AI bubble debate continues because neither side has a complete answer yet.

Bullish analysts can point to strong earnings, real demand for computing, and the possibility of large productivity gains.

Bearish analysts can point to massive capital expenditure, uncertain end demand, rising debt,t and concentrated market exposure.

Goldman Sachs Research occupies a middle position, by identifying bubble-like features while noting stronger corporate fundamentals than in previous speculative periods.

Jim Covello focuses on profitability.

Torsten Slok focuses on concentration.

BIS economists focus on financing and debt.

Jeff Buchbinder points to lower valuation extremes than the dot-com peak.

Michael Wilson emphasizes earnings growth.

These are not identical arguments.

They examine different parts of the same investment cycle.

For investors, the most useful approach is to track financial data rather than choosing a permanent bullish or bearish label.

AI revenue, capital expenditure, free cash flow, debt and valuation will determine whether current expectations prove realistic.

AurixFinance News will continue to monitor analyst disagreement surrounding AI investment, market concentration, and the financial returns generated by the global AI infrastructure build-out.

Financial Risk Notice

This article is for educational and informational purposes only. It does not constitute investment advice or a recommendation to buy or sell any security. AI-related investments can experience substantial price volatility. Investors should conduct independent research and consider their financial circumstances before making investment decisions.

About the Author

IFAZ Moshaddik, CFA

Market Strategist at AurixFinance News

IFAZ Moshaddik is a financial market analyst with more than 10 years of experience covering AI in finance, renewable energy stocks, and U.S. macroeconomics. A former Goldman Sachs analyst and CFA, he focuses his research on corporate earnings, capital expenditures, technology investment cycles, market valuations, and macroeconomic risk.

Sources

Published by: AurixFinance News

Website: www.aurixfinancial.com